localizer
trial_type coherent/down coherent/up incoherent/down incoherent/up
subject
01 30 30 30 30
02 30 30 30 30
03 30 30 30 30
05 30 30 30 30
06 30 30 30 30
07 30 30 30 30
08 30 30 30 30
09 30 30 30 30
10 30 30 30 30
11 30 30 30 30
12 30 30 30 30
Time course (Magnetometers)
Time course (Gradiometers)
Global field power
Time course (Magnetometers)
Time course (Gradiometers)
Global field power
Full-epochs Decoding
Based on N=10 subjects. Each dot represents the mean cross-validation score for a single subject. The dashed line is expected chance performance.
Decoding performance over time
Based on N=10 subjects. Standard error and confidence interval of the mean were bootstrapped with 5000 resamples. CI must not be used for statistical inference here, as it is not corrected for multiple testing. Time periods with decoding performance significantly above chance, if any, were derived with a one-tailed cluster-based permutation test (1023 permutations).
t-values based on decoding scores over time
Observed t-values. Time points with t-values > 1.833 were used to form clusters.
  """
hMT+ Localizer
"""
from mne_bids import get_entity_vals

study_name = 'ds003392'
bids_root = f'/storage/store2/data/{study_name}'
deriv_root = f'/storage/store2/derivatives/{study_name}/mne-bids-pipeline/'
subjects_dir = f'{bids_root}/derivatives/freesurfer/subjects'

# subjects = "all"
subjects = sorted(get_entity_vals(bids_root, entity_key='subject'))
exclude_subjects = ["06"]  # projs were applied during acquisition...
# subjects = sorted(list(set(subjects) - set(exclude_subjects)))
# subjects = ['01']
N_JOBS = len(subjects)
# N_JOBS = 1

task = 'localizer'
find_flat_channels_meg = True
find_noisy_channels_meg = True
use_maxwell_filter = True
ch_types = ['meg']

l_freq = 1.
h_freq = 40.
resample_sfreq = 250

# Artifact correction.
spatial_filter = 'ica'
ica_max_iterations = 500
ica_l_freq = 1.
ica_n_components = 0.99
ica_reject_components = 'auto'

# Epochs
epochs_tmin = -0.2
epochs_tmax = 1.0
baseline = (None, 0)

# Conditions / events to consider when epoching
conditions = ['coherent', 'incoherent']

# Decoding
decode = True
contrasts = [('incoherent', 'coherent')]

# Noise estimation
process_er = True
noise_cov = 'emptyroom'

on_error = "debug"

  Platform:         Linux-4.15.0-136-generic-x86_64-with-glibc2.27
Python:           3.9.9 | packaged by conda-forge | (main, Dec 20 2021, 02:41:03)  [GCC 9.4.0]
Executable:       /data/parietal/store/work/agramfor/mambaforge/bin/python3.9
CPU:              x86_64: 88 cores
Memory:           503.8 GB

mne:              1.1.1
numpy:            1.21.6 {blas=NO_ATLAS_INFO, lapack=lapack}
scipy:            1.8.1
matplotlib:       3.4.3 {backend=agg}

sklearn:          0.24.2
numba:            0.55.1
nibabel:          3.2.1
nilearn:          Not found
dipy:             Not found
cupy:             Not found
pandas:           1.3.3
pyvista:          0.32.1 {OpenGL 3.3 (Core Profile) Mesa 20.0.8 via llvmpipe (LLVM 10.0.0, 256 bits)}
pyvistaqt:        0.5.0
ipyvtklink:       Not found
vtk:              9.1.0
qtpy:             1.11.2 {PyQt5=5.12.9}
ipympl:           Not found
pyqtgraph:        Not found
pooch:            v1.6.0

mne_bids:         0.11.dev0
mne_nirs:         Not found
mne_features:     Not found
mne_qt_browser:   Not found
mne_connectivity: Not found
mne_icalabel:     Not found